Multi-Agent Platform
Multi-agent workflows in production, modular across LLM providers.
- Agentic AI
- RAG
- Multi-LLM
- GCP
- dbt
- Responsible AI
Problem
Enterprise customers needed AI that worked reliably in production rather than as demos, without being locked to a single model provider.
Problem
Enterprise customers needed AI that worked reliably in production rather than as demos, without being locked to a single model provider.
What I built
Multi-agent workflows in a provider-modular architecture spanning Claude, Gemini and other LLMs, plus local open-source models, with RAG pipelines and task-specific agents. Architected end to end, from model selection to production Python and dbt data transformations, on a Google Cloud stack.
Select a part of the diagram to see what it does.
Outcome
Established responsible-AI governance and quality practices for reliable, scalable deployment, and advanced customers' AI strategy from concept through proof-of-value to production.